Water quality monitoring using a smart sensing system

This paper presents an innovative smart sensor used for water quality monitoring. This smart sensing system utilizes spectroscopic techniques combined with the measurement of physico-chemical variables, to estimate global pollution parameters in water samples, particularly the Chemical Oxygen Demand (COD). This estimation is computed using a multisensor fusion approach, by means of an artificial neural network algorithm. The smart sensor has been tested successfully over a set of 71 wastewater samples of the city Pierre Benite (France) and it was shown that the estimated values of COD were in good agreement with the observed COD values, which were measured with a conventional method. (C) 2000 Elsevier Science Ltd. All rights reserved.